Anakonda's profile - activity

It has been some time since my Sage stopped working properly. It seems symbolic computations needing libecl are broken:

sage: var('x')
x
sage: solve([x],x)

Internal or unrecoverable error in:
Got signal before environment was installed on our thread
  [2: No such file or directory]

;;; ECL C Backtrace
;;; /usr/lib64/libecl.so.13.5(si_dump_c_backtrace+0x31) [0x7f30c79e45f1]
;;; /usr/lib64/libecl.so.13.5(ecl_internal_error+0x44) [0x7f30c79cd5c4]
;;; /usr/lib64/libecl.so.13.5(+0x1b8578) [0x7f30c79fa578]
;;; /lib64/libpthread.so.0(+0xfeb0) [0x7f3105cc5eb0]
;;; /usr/lib64/libgc.so.1(+0x8bcd) [0x7f30f2130bcd]
;;; /usr/lib64/libgc.so.1(+0x68eb) [0x7f30f212e8eb]
;;; /usr/lib64/libgc.so.1(+0x6c4e) [0x7f30f212ec4e]
;;; /usr/lib64/libgc.so.1(+0x13a94) [0x7f30f213ba94]
;;; /usr/lib64/libgc.so.1(+0x87d3) [0x7f30f21307d3]
;;; /usr/lib64/libgc.so.1(+0xd659) [0x7f30f2135659]
;;; /usr/lib64/libgc.so.1(GC_generic_malloc+0x47) [0x7f30f2135757]
;;; /usr/lib64/libgc.so.1(+0xdada) [0x7f30f2135ada]
;;; /usr/lib64/libecl.so.13.5(ecl_alloc_object+0x9b) [0x7f30c7a10f7b]
;;; /usr/lib64/libecl.so.13.5(init_threads+0x42) [0x7f30c7a0e172]
;;; /usr/lib64/libecl.so.13.5(cl_boot+0x87) [0x7f30c78dd377]
;;; /usr/lib64/python2.7/site-packages/sage/libs/ecl.so(+0x6c6c) [0x7f30c7d6dc6c]
;;; /usr/lib64/python2.7/site-packages/sage/libs/ecl.so(+0x6f98) [0x7f30c7d6df98]
;;; /usr/lib64/python2.7/site-packages/sage/libs/ecl.so(initecl+0x1781) [0x7f30c7d79961]
;;; /usr/lib64/libpython2.7.so.1.0(_PyImport_LoadDynamicModule+0x99) [0x7f3105fc4a39]
;;; /usr/lib64/libpython2.7.so.1.0(+0xefc29) [0x7f3105fc2c29]
;;; /usr/lib64/libpython2.7.so.1.0(+0xefe6d) [0x7f3105fc2e6d]
;;; /usr/lib64/libpython2.7.so.1.0(PyImport_ImportModuleLevel+0x188) [0x7f3105fc37d8]
;;; /usr/lib64/libpython2.7.so.1.0(+0xd792f) [0x7f3105faa92f]
;;; /usr/lib64/libpython2.7.so.1.0(PyObject_Call+0x43) [0x7f3105f1d593]
;;; /usr/lib64/libpython2.7.so.1.0(PyEval_CallObjectWithKeywords+0x47) [0x7f3105fac477]
;;; /usr/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0xf05) [0x7f3105fad965]
;;; /usr/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7dd) [0x7f3105fb251d]
;;; /usr/lib64/libpython2.7.so.1.0(PyEval_EvalCode+0x32) [0x7f3105fb2622]
;;; /usr/lib64/libpython2.7.so.1.0(PyImport_ExecCodeModuleEx+0x8c) [0x7f3105fc1d9c]
;;; /usr/lib64/libpython2.7.so.1.0(+0xef005) [0x7f3105fc2005]
;;; /usr/lib64/libpython2.7.so.1.0(+0xefc29) [0x7f3105fc2c29]
;;; /usr/lib64/libpython2.7.so.1.0(+0xefe6d) [0x7f3105fc2e6d]
Aborted

It also happens when I try to use the sum function with symbolic variables etc. The problem is not in ECLS per se, since maxima solves these things just fine.

I'm running Sage 6.5 on Gentoo Linux (using the sage-on-gentoo overlay), but the errors have been there a long time (I think < 6.2). I did various reinstalls of all my packages, also completely purged libecls, maxima, Sage and reinstalled them, but to no effect. I also tried installing all the relevant packages without threads support, as well as different versions of ECLS, nothing changed.

Does anyone have an idea what could cause this problem? Or what is even meant by the error? What file or directory isn't found?

I have the same problem now (iterating over some polynomials, the memory for the groebner base is not freed). I'm thinking about filing a bug report. Did you find any solution?

I think maybe this works. I'm not an mpl expert, but at least the points correspond to the output of p.

sage: var('y')
y
sage: p = implicit_plot(x^2+y^2-1,(x,-2,2),(y,-2,2),plot_points=4)
sage: P = p.matplotlib()
sage: R = P.get_children()[1]
sage: S = R.collections[0]
sage: S.get_paths()
[Path([[-0.66666667 -0.70833333]
 [-0.70833333 -0.66666667]
 [-0.70833333  0.66666667]
 [-0.66666667  0.70833333]
 [ 0.66666667  0.70833333]
 [ 0.70833333  0.66666667]
 [ 0.70833333 -0.66666667]
 [ 0.66666667 -0.70833333]
 [-0.66666667 -0.70833333]], None)]
sage: S = R.collections[1]
sage: S.get_paths()
[Path([[-0.66666667 -0.70833333]
 [-0.70833333 -0.66666667]
 [-0.70833333  0.66666667]
 [-0.66666667  0.70833333]
 [ 0.66666667  0.70833333]
 [ 0.70833333  0.66666667]
 [ 0.70833333 -0.66666667]
 [ 0.66666667 -0.70833333]
 [-0.66666667 -0.70833333]], None)]

Yes, this does exactly what I wanted. This is just great. I never would have guessed that descent into the depths of the matplotlib arrays. Thank you very much!

I'm looking for a way to get a plain list of coordinates from a plot object. For instance, after creating an implicit plot of a circle

p = implicit_plot(x^2+y^2-1,(x,-2,2),(y,-2,2))

I am able to save the image in .pdf or even .svg format (where coordinates are present), which is a very good start. What I really want, though, is a plain list of coordinates of, say, 2000 points lying on that circle. It should be possible to extract some from the .svg file, but is there a more convenient method? Maybe even one where I can set the sample size?

Thank you in advance for any suggestions.